1. Field of the Invention
The present invention relates to an electrostatic capacitive sensor and a method for producing the same sensor.
2. Description of the Related Art
Progress is rapidly being made in silicon micromachining technology in which various types of miniature parts are produced by using a very-high-precision etching method for single-crystal silicon and a polysilicon-deposition method, such as chemical vapor deposition (CVD) or physical vapor deposition (PVD). As a consequence, various types of miniature parts, such as electrostatic capacitive sensors for detecting, for example, acceleration and angular velocity, have been developed.
The principle of the capacitive sensor will now be explained with reference to FIG. 5A. The capacitive sensor is constructed of a sensing unit 1 and a CV conversion circuit 2 for converting capacitance into voltage.
The sensing unit 1 is formed of a movable electrode 3 and a stationary electrode 4, both of which are flat-plate type and placed parallel to each other with their surfaces opposedly facing. When acceleration is applied perpendicular to the surface of the movable electrode 3, the electrode 3 is displaced in a direction indicated by the arrow D1 shown in FIG. 5A to change the distance L1 between the movable electrode 3 and the stationary electrode 4. Accordingly, the capacitance between the two electrodes 3 and 4 is changed in proportion to the reciprocal of the distance L1 therebetween. In contrast, upon application of acceleration parallel to the surface of the movable electrode 3, the electrode 3 is displaced in a direction indicated by the arrow D2 shown in FIG. 5A to vary the opposing area of the electrodes 3 and 4. Consequently, the capacitance between the electrodes 3 and 4 is changed in proportion to the opposing area.
As the CV conversion circuit 2, a capacitance detecting circuit using what is referred to as a "diode bridge" is generally employed. The input terminals of the conversion circuit 2 are electrically connected to the movable electrode 3 and the stationary electrode 4, respectively, via individual lead lines 5. As a consequence, a change in the capacitance between the electrodes 3 and 4 is converted into a voltage by the CV conversion circuit 2, thereby detecting the acceleration.
In the principle explained above, the sensing unit 1 has been provided by way of example only, and it is not restricted to the above-described type. For example, the sensing unit 1 may be formed, as illustrated in FIG. 5B, of a movable electrode 3 and a pair of stationary electrodes 4A and 4B. In this case, the electrodes 4A and 4B are disposed parallel to each other with their surfaces opposedly facing. Further, the movable electrode 3 is located parallel to and between the stationary electrodes 4A and 4B in such a manner that the top and bottom surfaces of the electrode 3 opposedly face the electrodes 4A and 4B. The input terminals of the CV conversion circuit 2 are electrically connected to the movable electrode 3 and the stationary electrodes 4A and 4B, respectively, via individual lead lines 5. Upon application of acceleration perpendicular to the surface of the movable electrode 3, the electrode 3 is displaced in a direction indicated by the arrow D1 shown in FIG. 5B to change the distance L2 between the movable electrode 3 and the stationary electrode 4A and the distance L3 between the electrodes 3 and 4B. Accordingly, changes in the capacitance between the movable electrode 3 and the stationary electrode 4A and the capacitance between the electrodes 3 and 4B are converted into a differential voltage by the CV conversion circuit 2, thereby detecting the acceleration. It should be noted that an explanation of the structure of the differential-voltage-type sensor will be omitted in a specific example of the capacitive sensor to be described below.
A typical known type of electrostatic capacitive sensor will now be explained more specifically with reference to FIGS. 6A and 6B.
The capacitive sensor is constructed of a support base 6, a sensing unit 7, and a CV conversion circuit 8. The support base 6 is formed in the shape of a quadrilateral plate and made of, for example, electrically insulating glass. A rectangular recessed portion (opening) 9 is formed substantially at the center of the surface of the support base 6, while another rectangular recessed portion (opening) 10 is provided in the vicinity of one edge of the base 6. The recessed portions 9 and 10 are disposed parallel to each other.
The sensing unit 7 is formed of a stationary electrode 11 and a movable electrode 12, both of which exhibit electrical conductivity. The sensing unit 7 is produced by using single-crystal silicon doped with impurity ions, such as phosphorus, boron, or antimony. The stationary electrode 11, being formed in the shape of a rectangular plate, is provided projecting from a longitudinal edge of the recessed portion 9 on the opposite side thereof, away from the other recessed portion 10 formed in the support base 6.
The movable electrode 12 is integrally constructed of a pair of support portions 13, a mass portion 14, and a pair of interconnecting portions 15 for connecting the mass portion 14 to the respective support portions 13.
The pair of support portions 13 are formed in the shape of a quadrangular prism and project from the respective end edges of the recessed portion 9. The support portions 13 are disposed with their major side faces opposedly facing each other. The height of the support portions 13 is the same as the stationary electrode 11.
The mass portion 14 is formed in the shape of a rectangular prism and is interposed between the opposing side faces of the support portions 13. The mass portion 14 is disposed parallel to the stationary electrode 11 with a predetermined spacing in such a manner that one longitudinal face of the mass portion 14 opposedly faces one longitudinal face of the stationary electrode 11. The height of the mass portion 14 is the same as the stationary electrode 11.
The interconnecting portions 15 are formed in the shape of a thin rectangular plate and each connect the end face of the mass portion 14 to the opposing face of the support portion 13. The interconnecting portion 15 is identical to the mass portion 14 in height and perpendicularly provided at the center of the end face of the mass portion 14. With this arrangement, the mass portion 14 is held by the interconnecting portions 15 in such a manner that it floats over the recessed portion 9. The interconnecting portions 15 are formed thin in a direction perpendicular to the longitudinal face of the mass portion 14 so that they can be easily deformed in a bending manner in the same direction. Thus, when acceleration is applied perpendicular to the longitudinal faces of the mass portion 14, the interconnecting portions 15 are deformed in a bending manner in the direction in which acceleration is applied, thereby changing the distance between the mass portion 14 and the stationary electrode 11.
The CV conversion circuit 8 is formed at the center of the bottom surface of a single-crystal-silicon block 16. The block 16 is located on the recessed portion 10 in such a manner that the peripheral edge on the bottom surface of the block 16 is disposed around the recessed portion 10. Accordingly, the CV conversion circuit 8 is held at the center of the recessed portion 10 without directly contacting the insulating substrate 6. The height of the block 16 is the same as the stationary electrode 11. The input terminals of the CV conversion circuit 8 are electrically connected to the stationary electrode 11 and the movable electrode 12, respectively, via individual lead lines 17 formed on the top surface of the support base 6 and the bottom surface of the block 16. With this construction, a change in capacitance between the mass portion 14 and the stationary electrode 11 can be converted into a voltage by the CV conversion circuit 8, thereby detecting the acceleration applied to the mass portion 14 of the movable electrode 12.
A brief explanation will now be given of a manufacturing method for the electrostatic capacitive sensor constructed as described above while referring to FIGS. 7A through 7E.
Silicon nitride (SiNx) film 19 is deposited, as illustrated in FIG. 7A, on the top and bottom surfaces of a single-crystal silicon substrate 18 by using the low-pressure CVD method. Further, the silicon nitride film 19 on the bottom surface of the substrate 18 is patterned into a predetermined shape using photolithographic technology and anisotropic etching technology, thereby forming etching masks 19A, 19B and 19C.
Subsequently, the single-crystal silicon substrate 18 is allowed to react with an alkaline aqueous solution, such as potassium hydroxide (KOH), so as to perform anisotropic-etching on the portions not covered by the etching masks 19A through 19C. As a result, a plurality of projecting portions 18A, 18B and 18C are formed, as shown in FIG. 7B. As will be described later, the projecting portions 18A, 18B and 18C correspond to the stationary electrode 11, and the mass portion 14 of the movable electrode 12, and the CV conversion circuit 8, respectively. It should be noted that the projecting portions 18A, 18B and 18C are integrally formed with the remaining silicon substrate 18 protected from etching.
The etching masks 19A through 19C are then removed, as shown in FIG. 7C, by means such as chemical etching using phosphoric acid or reactive ion etching (RIE). Thereafter, the projecting portions 18A and 18B are doped with impurity ions using techniques such as thermal diffusion or ion implantation, thereby making the projecting portions 18A and 18B electrically conductive. Moreover, the CV conversion circuit 8 is formed at the center of the forward end face of the projecting portion 18C according to a known semiconductor-integrated-circuit formation technique.
Then, the forward end faces of the projecting portions 18A, 18B and 18C are overlaid, as illustrated in FIG. 7D, on the surface of the support base 6 provided with the recessed portions 9 and 10 by means such as RIE. More specifically, the forward end face of the projecting portion 18A is placed at the edge of the recessed portion 9, while the forward end face of the projecting portion 18B is positioned at the center of the recessed portion 9. Further, the peripheral edge of the forward end face of the projecting portion 18C is disposed on the recessed portion 10 while contacting the edge of the portion 10. The forward end faces of the projecting portions 18A, 18B and 18C are then fixed on the support base 6 by means such as anode coupling or fusion coupling.
Subsequently, the remaining single-crystal silicon substrate 18 protected from etching is removed, as illustrated in FIG. 7E, using techniques such as RIE or polishing, thereby separating the projecting portions 18A, 18B and 18C. As a consequence, the stationary electrode 11, the movable electrode 12, and the block 16 are formed, thus completing an electrostatic capacitive sensor.
In the aforedescribed capacitive sensor, since the sensing unit 7 is formed by etching the single-crystal silicon substrate 18, the stationary electrode 11 and the mass portion 14 of the movable electrode 12 can be made large. This increases the opposing area of the stationary electrode 11 and the mass portion 14, thereby achieving a large capacitance therebetween.
However, in the above-described electrostatic capacitive sensor, since anode coupling or fusion coupling is performed to fix the substrate provided with the projections onto the support base, high voltages or high heat used during the substrate-mounting operation may produce an adverse influence on the CV conversion circuit. This may cause a breakdown of the circuit.
Moreover, a recessed portion is provided for the support base to keep the CV conversion circuit from directly contacting the support base in order to relax a distortional stress generated therebetween. However, a distortional stress produced between the peripheral edge of the block and the support base may sometimes adversely influence the CV conversion circuit, causing the circuit to break. To overcome this drawback, one of the measures to be taken may be to form a CV conversion circuit on the top surface of the block which is less vulnerable to a distortional stress. However, this measure requires the provision of lead lines on the lateral surfaces as well as on the top surface of the block to electrically connect the CV conversion circuit with the stationary electrode and the movable electrode, which makes the provision of the lead lines difficult. Additionally, if the stationary electrode and the CV conversion circuit are integrally formed, the area of contact between these elements and the support base is unfavorably increased to generate a larger distortional stress, which adversely affects the CV conversion circuit more severely. Accordingly, the stationary electrode and the CV conversion circuit are required to be separately formed, thereby making the manufacturing process complicated and also enlarging the resulting sensor.